Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Tested Monstro vs. Helium

Let's hope the test will be done by someone who is highly respected in the film business and is not (when this is even possible) biased.

Deakins, Van Hoytema, Yedlin, Fincher,....
 
Proper test excludes the bias.

No need for superstars. Plenty of skilled DPs here to do it.

Idolatry propagates ideology and has no place in science.
 
One thing I'll add about differences btw Venice, Monstro and Alexa LF is...how flexible 8K Monstro can be vs 6K Venice, or 4K Alexa LF.

Even if Venice or LF had a stop better DR, that would still be far less useful for my team...than how versatile the 8K Monstro files would be.

Lemme explain.

We currently shoot a lot of 6K Dragon, and one of the features our clients love is the 4K extraction we can pull from this. Almost every interview we shoot we deliver graded 1080 quicktimes of both the 6K master, and the 4K reframe, and our broadcast clients just LOVE this. It's helps them A LOT. Especially on fast turn-arounds where that extra punch helps them cut down bites faster, and knock out edits quicker.

Here's some recent examples.

Here's a spot that only used the 4K reframe:



And here's a spot that uses both the 6K master and 4K reframe:



What exciting for me about Monstro is the idea of geting an ADDITIONAL reframe. That is, we could shoot an 8K wide, reframe for a 6K Medium, and then reframe once more for a 4K Closeup.

Three shots in one is kind of amazing.


And something that neither Venice nor LF could as easily do.

Just something to think about.

Respecfully,
 
Last edited:
Arri has beaten out the competition for many reasons but one of the main ones is the color science, especially in regards to skintones in whatever lighting scenario. IPP2 is a massive improvement for RED but it still doesn't have the same cinematic quality as Alexa footage, even with this new Monstro sensor. If RED finally makes a color profile that looks like film or is more cinematic in general, they will regain market share. But until that happens, film and television will still be Arri dominated.

Why won't RED hire a company like Light Iron to come up with a color science that colorists and DP's actually want? I know the RED fanboys will say they love Red's color but why isn't this camera being used for the majority or acclaimed movies and T.V.? Surely improving on the color would bridge the gap, especially with the clean dynamic range the monstro sensor has.
 
Arri has beaten out the competition for many reasons but one of the main ones is the color science, especially in regards to skintones in whatever lighting scenario. IPP2 is a massive improvement for RED but it still doesn't have the same cinematic quality as Alexa footage, even with this new Monstro sensor. If RED finally makes a color profile that looks like film or is more cinematic in general, they will regain market share. But until that happens, film and television will still be Arri dominated.

Why won't RED hire a company like Light Iron to come up with a color science that colorists and DP's actually want? I know the RED fanboys will say they love Red's color but why isn't this camera being used for the majority or acclaimed movies and T.V.? Surely improving on the color would bridge the gap, especially with the clean dynamic range the monstro sensor has.

Couldn't agree more, Arri has won DP's over because of the very simple fact you get a 5213 colored image straight in the card, whereas with red you need to work a lot in the grade to get it there, because as opposed to recording an aesthetically pleasing image, it leans way more towards a "scientifically accurate" color reproduction. If they re-design the color you get on the card, I think Red will instantly win back some market share.
 
I think Helium with standard OLPF looks really good for skintones per Phil Holland's test : https://www.youtube.com/watch?v=CvtC1Kje4EQ
I think most DPs don't use red because they are used to the problems that came with the dragon with skintone highlight's greenish cast to skintones.

Helium is a really, really good sensor and I think has a chance of being more popular. And Monstro looks to improve on the helium sensor even more.
 
I don't know if Red have a particular ISO rating that they claim as 'native' for Monstro, but it just occurred to me that Panavision call it 1600 ISO for the base on their new DXL2, which means that at 2000 ISO (as in this test) the camera is only 1/3 stop underexposed. Which would certainly help explain how incredibly clean the image is compared to Venice (which is two stops under here), and Helium (which is 1 + 1/3 under).
 
Couldn't agree more, Arri has won DP's over because of the very simple fact you get a 5213 colored image straight in the card, whereas with red you need to work a lot in the grade to get it there, because as opposed to recording an aesthetically pleasing image, it leans way more towards a "scientifically accurate" color reproduction.

This keeps being repeated but it is incorrect.

Arri image has some aspects more accurate. Rad image has some aspects more accurate. One cannot generally say that one is more or less accurate than the other. What you like in Arri image is due to some things being more accurate.
 
Why won't RED hire a company like Light Iron to come up with a color science that colorists and DP's actually want?

Greetings.

In particular the ghostStocks from philmColor.

http://www.philmcolor.com

And if you are super curious, you'll want to stay tuned for what is coming in Release 3.

I've been working rather diligently on this particular endeavor for a long while now. philmColor has been used by Netflix, Warner Brothers, and several other studios at this point. Give it a look.
 
Greetings.

In particular the ghostStocks from philmColor.

http://www.philmcolor.com

And if you are super curious, you'll want to stay tuned for what is coming in Release 3.

I've been working rather diligently on this particular endeavor for a long while now. philmColor has been used by Netflix, Warner Brothers, and several other studios at this point. Give it a look.

Hey Phil,
I think I am ready to pull the trigger for philcolor. I remember you once had this discount coupon. Is it still valid?
I really like the colors from IPP2 and Helium but would love to give that extra touch to my footage without having to go through color grading, especially for quick turn around projects (not talking about color corrections and matching which is a different thing though).

Btw, can you point out which Netflix and WB series did use philcolor? I am curious.

Cheers,
Thai
 
Hey Phil,
I think I am ready to pull the trigger for philcolor. I remember you once had this discount coupon. Is it still valid?
I really like the colors from IPP2 and Helium but would love to give that extra touch to my footage without having to go through color grading, especially for quick turn around projects (not talking about color corrections and matching which is a different thing though).

Btw, can you point out which Netflix and WB series did use philcolor? I am curious.

Yep, the code is "phfxlovesyou".

There's a few :) But stay tuned on that front.
 
This keeps being repeated but it is incorrect.

Arri image has some aspects more accurate. Rad image has some aspects more accurate. One cannot generally say that one is more or less accurate than the other. What you like in Arri image is due to some things being more accurate.

Having shot both Arri and my own Dragon of course, I've always found Red to be closer to what my eye sees in a scene, with exceptions of course, whereas with the Alexa, even through the monitor I can see clearly it's reproducing some colors, particularly blues and greens, like a Kodak stock, so what I mean by "accurate" is that Red leans more towards colors that are the same to what my eye sees whereas Arri gives a more Kodak-like reproduction of tones on the card.
 
Having shot both Arri and my own Dragon of course, I've always found Red to be closer to what my eye sees in a scene, with exceptions of course, whereas with the Alexa, even through the monitor I can see clearly it's reproducing some colors, particularly blues and greens, like a Kodak stock, so what I mean by "accurate" is that Red leans more towards colors that are the same to what my eye sees whereas Arri gives a more Kodak-like reproduction of tones on the card.
Colors is a matter of taste and preference.
A good colorist can make any color out of any camera.
Look at some blockbusters, they shoot some scenes with multicam from Arri, Canon 5D, and Gopros, and you won‘t notice a thing.
 
Having shot both Arri and my own Dragon of course, I've always found Red to be closer to what my eye sees in a scene, with exceptions of course, whereas with the Alexa, even through the monitor I can see clearly it's reproducing some colors, particularly blues and greens, like a Kodak stock, so what I mean by "accurate" is that Red leans more towards colors that are the same to what my eye sees whereas Arri gives a more Kodak-like reproduction of tones on the card.

Yes for sure Arri is more trying to mimic film emulsions than the looking out of the window look kind of look. But to me thats a rather strange strive for their side. I think its strange that they dont push more luts that is more like "clear glas".

Red got the more scientific approach and to me thats more appealing creating a look should not just come included with the camera. Better to have such as something you add, choose to put on.

But yes, both black magic and arri to me have more of a film look baked that is quite difficult to get around. Also Kinefinity got the desaturated a bit bleach bypassed look, looks great i think but rather go for that with luts or in post.
 
The problem I have with RED is the skin tones. They slant orange and they’re either too saturated or not saturated enough for certain hues. This is not as much of an issue every time I’ve used an Alexa. Arri somehow allows hues outside of skin to be saturated but somehow always manages to keep skin tones in the sweet spot of saturation with nuanced hues even in mixed lighting scenarios.

My RED cameras always look good in daylight but the minute there’s a mixed lighting scenario the colors are hard to RE balance compared to an Alexa and, sadly, no 3rd party LUT has been able to fix this. They’ve all improved on some of the problems I have with RED footage but they can never seem to solve this one underlying problem. Shout out to Truecololr and Omeneo LUTs for maxing out on IPP2. Haven’t tried Philcolor yet.

At the moment, I have some pretty crazy nodes going in Resolve to get an image I like and that involves qualifying all the major colors and skin in a parallel mixer node so I can tweak the color especially with the sat vs sat curve (this especially helps reduce the dreaded green tinge to Helium footage. But this requires so much unnecessary work in my opinion. I recently edited and colored an Alexa project and was so impressed by not having to do this for the footage. It was easy to work with Right off the bat (but I was surprised at how noisy it is when you view it on a huge 4k monitor, especially the blue channel).

I played with some Monstro clips that are on Reduser and the sensor is a definitely the best one RED has made but it still does not completely solve the above issues. This needs to be fixed on a color science level.
 
The problem I have with RED is the skin tones. They slant orange and they’re either too saturated or not saturated enough for certain hues. This is not as much of an issue every time I’ve used an Alexa. Arri somehow allows hues outside of skin to be saturated but somehow always manages to keep skin tones in the sweet spot of saturation with nuanced hues even in mixed lighting scenarios.

My RED cameras always look good in daylight but the minute there’s a mixed lighting scenario the colors are hard to RE balance compared to an Alexa and, sadly, no 3rd party LUT has been able to fix this. They’ve all improved on some of the problems I have with RED footage but they can never seem to solve this one underlying problem. Shout out to Truecololr and Omeneo LUTs for maxing out on IPP2. Haven’t tried Philcolor yet.

At the moment, I have some pretty crazy nodes going in Resolve to get an image I like and that involves qualifying all the major colors and skin in a parallel mixer node so I can tweak the color especially with the sat vs sat curve (this especially helps reduce the dreaded green tinge to Helium footage. But this requires so much unnecessary work in my opinion. I recently edited and colored an Alexa project and was so impressed by not having to do this for the footage. It was easy to work with Right off the bat (but I was surprised at how noisy it is when you view it on a huge 4k monitor, especially the blue channel).

I played with some Monstro clips that are on Reduser and the sensor is a definitely the best one RED has made but it still does not completely solve the above issues. This needs to be fixed on a color science level.

Arri clamp everything orange / red towards nice skin tones. Just as they clamp blue towards more cyan blue sky kind of blue. Just as Kodak does. Possibly it looks good, I think so, but it should not be part of the camera but the grading.

Top clamped colors bottom straight colors / correct colors, still I find the top one to look better but it´s not right.

PastedGraphic-17 by Björn Benckert, on Flickr
 
The problem I have with RED is the skin tones. They slant orange and they’re either too saturated or not saturated enough for certain hues. This is not as much of an issue every time I’ve used an Alexa. Arri somehow allows hues outside of skin to be saturated but somehow always manages to keep skin tones in the sweet spot of saturation with nuanced hues even in mixed lighting scenarios.

My RED cameras always look good in daylight but the minute there’s a mixed lighting scenario the colors are hard to RE balance compared to an Alexa and, sadly, no 3rd party LUT has been able to fix this. They’ve all improved on some of the problems I have with RED footage but they can never seem to solve this one underlying problem. Shout out to Truecololr and Omeneo LUTs for maxing out on IPP2. Haven’t tried Philcolor yet.

At the moment, I have some pretty crazy nodes going in Resolve to get an image I like and that involves qualifying all the major colors and skin in a parallel mixer node so I can tweak the color especially with the sat vs sat curve (this especially helps reduce the dreaded green tinge to Helium footage. But this requires so much unnecessary work in my opinion. I recently edited and colored an Alexa project and was so impressed by not having to do this for the footage. It was easy to work with Right off the bat (but I was surprised at how noisy it is when you view it on a huge 4k monitor, especially the blue channel).

I played with some Monstro clips that are on Reduser and the sensor is a definitely the best one RED has made but it still does not completely solve the above issues. This needs to be fixed on a color science level.


Anton,


I myself have made up a few funky node trees to try to compensate for the greenish skin tones. So I'm currently trying an alternative method in 3DLUT Creator Pro 1.46


1) First I exported a RedwidegamutRGB/Log3G10 Tiff Out of Redcine-X Pro with a "5000" Temp and "-16.00" tint.


2) Next I open up "3DLut Creator Pro" and under "File" chose "Open Image", which brought the Rwg/Lg3g10 .Tiff in.


40932683852_6153d05f44_b.jpg
[/url]Screenshot (3) by rand thompson, on Flickr[/IMG]



3) In the "Channels" Tab, Select the "down arrow" at the right and choose "Create Color Space Transform Matrix"


40932683452_691a7d05e6_b.jpg
[/url]Screenshot (4) by rand thompson, on Flickr[/IMG]


4) Under "Input Primaries" choose "Red RedWideGamut" and under "Output Primaries" choose "sRGB/ Rec709"


40265114184_0c150bfdb0_b.jpg
[/url]Screenshot (9) by rand thompson, on Flickr[/IMG]



39165838000_2dfd65b61e_b.jpg
[/url]Screenshot (28) by rand thompson, on Flickr[/IMG]




5) Here is what that looks like. Some of the numbers have been "rounded up" but I'm still trying to see how much of a difference that makes.


27103814618_d2602b1fe8_b.jpg
[/url]Screenshot (24) by rand thompson, on Flickr[/IMG]



26103167727_76ebeb2879_b.jpg
[/url]Screenshot (18) by rand thompson, on Flickr[/IMG]



6) Here's what the image and the Rwg/Log3G10 looks like represented. You will still have to create your own luminance curve however.


40932682982_c100ba1bf9_b.jpg
[/url]Screenshot (8) by rand thompson, on Flickr[/IMG]


7) Here's The "Master Luminance curve I created plus the image. Skin is still a little orange.

40265113614_1edce9cf74_b.jpg
[/url]Screenshot (13) by rand thompson, on Flickr[/IMG]

26103167707_7838fb254a_b.jpg
[/url]Screenshot (19) by rand thompson, on Flickr[/IMG]



8) Compared it to a "Medium Contrast" "Soft Roll-Off" exported image(Top) as "Reference" out of Redcine-X to try to approximate the contrast curve to modify later.


40265113874_bea1a5d189_b.jpg
[/url]Screenshot (12) by rand thompson, on Flickr[/IMG]


9) Then I went to the "A/B' colorwheel to try to work with skin tones.


27103814798_6581f36783_b.jpg
[/url]Screenshot (23) by rand thompson, on Flickr[/IMG]



10) Here's a graphical representation of what the curve and "A/B" correction did so far.

40272244484_0ee9dc7913_b.jpg
[/url]Screenshot (14) by rand thompson, on Flickr[/IMG]


Still just F**king around however until I understand a whole lot about this program and the color transform mathematics.
 
Last edited:
Rand this is very interesting indeed. I've never used 3D LUT creator and I'm now definitely going to check it out. I just wish that it was something easy to do inside of Resolve. This looks looks like one more step removed even though in theory you only have to do it once. However, in my experience dealing with different white balance scenarios the offsets I use don't always work consistently, you often have to retweak on a case to case basis.

Bjorn, the top image looks better to me to because there's more color contrast. The lower one looks like it has an overall orange/greenish tint.

When I tried to make an Alexa to RED LUT myself about a year ago, based on the color charts that people made available on REDUSER, I found that a lot of the difference between RED and Alexa is the way the cameras map luminance towards certain colors at certain exposures. For example, Alexas not only map blue more Cyan but they also map it darker than RED giving it a richer tone. I also found that Alexa clamps colors based on saturation level and luminance level meaning they use some kind of highpass filter that allows saturated colors to go through but desaturate colors below a certain level (hece Alexas have less color cast issues than RED cameras). The same goes for the luminance levels of colors. Alexa definitely desaturates colors above and below a certain luminance threshold. And, as Bjorn mentioned earlier, they especially do this for any colors close to skintones. This is where the Alexa really shines. Skintones are not easy to get perfect and the Alexa color science makes getting skin right a lot easier than any other camera out there.

I sincerely believe that RED has the sensor tech to make amazing colors but something is getting lost in translation. LUT's can fix certain aspects of an image but they definitely are not a cure all. The luminance mapping of colors is especially hard to get right in a one-LUT-fits all. This has to be done in a camera at a color science level. Panavision recognized this issue and that's why they got Light Iron involved. Now RED just has to do the same.
 
Another theory I have about the Alexa color science has to do with its pixel size. The bigger the photosite, the less crosstalk which means more accurate colors. This, coupled with its dual gain sensor means its combining 2 readouts at optimum levels. In other words, the Alexa is the OG HDR cinema sensor. Its exposing for the shadows and the highlights and then combining the image giving an image that has colors accurate throughout its exposure range.

This camera tech, however, comes with a cost which is resolution, heat, and power consumption. I talked to an abelcine rep in New York when they had the new Alexa Full Frame camera on display and he told me there's no Alexa mini 2 yet because of heat issues. That dual readout comes at a cost. He did, however, tell me that they're would be a new mini in 1-2 years so apparently they're figuring it out...
 
Back
Top